The RATS Software Forum

Posted: **Fri Sep 09, 2011 5:58 am**

Hi,

The paper replication example, Perron-Wada (http://www.estima.com/forum/viewtopic.php?f=8&t=1504), includes an estimation of a SS-model with a broken trend rae. Is there a clever way to introduce more than one break in the trend rate?. I tried but unfortunately only with embarassing results.

Thanx

Marcus

Posted: **Fri Sep 09, 2011 9:35 am**

That's fairly straightforward. This does breaks at 1973:1 (as in the paper) plus another at 1990:1. The three lines with >>> are the only ones that need to change.

Code: Select all

set lgdp  = 100.0*lgdp
set d1973 = t>1973:1
>>>SET D1990 = T>1990:1
*
@NBERCycles(down=recession)
*
* UC-0 decomposition with AR(2) cycle, broken trend rate.
*
>>>nonlin mu D1 D2 sn ph1 ph2 se
*
dec frml[rect] af
dec frml[vect] zf
dec frml[symm] swf
*
frml af = ||1.0,0.0,0.0|$
            0.0,ph1,ph2|$
            0.0,1.0,0.0||
>>>frml zf = ||mu+D1*D1973+D2*D1990,0.0,0.0||

The RATS Software Forum

More than one broken trend rate

More than one broken trend rate

Re: More than one broken trend rate